Coronary artery disease (or CAD) is the number one cause of death in the United States (Nature Med 2002, 8:1209-1262). The initiation and progression of CAD involves a complex interplay between multiple physiological processes, including inflammation, lipid homeostasis, and insulin resistance/diabetes mellitus. Multiple clinical studies have now shown that the three primary components of plasma lipids, low-density lipoprotein (or LDL), high-density lipoproteins (or HDL), and triglycerides (or TGs), are causally associated with the propensity to develop atherosclerosis and CAD. Along side other risk factors such as positive family history of CAD, elevated body-mass index, hypertension, and insulin resistance/diabetes mellitus, elevated plasma LDL and/or TG-rich lipoproteins and decreased plasma HDL levels have been defined as major cardiovascular risk factors by the National Cholesterol Education Program Adult Treatment Panel 111 (NCEP ATP III; Am J Cardio 2003, 92: 19i-26i). Accordingly, therapeutic intervention strategies designed to impact these plasma lipid components as well as those that underlie insulin resistance are of great interest to the medical community.
In terms of LDL-lowering, drugs of the statin class are structurally similar to the molecule hydroxymethylglutaryl-coenzyme A (HMG-CoA), a biosynthetic precursor of cholesterol. These drugs are competitive inhibitors of the rate-limiting step of cholesterol biosynthesis catalyzed by HMG-CoA reductase. Mechanistically, the statins lower LDL by upregulating the LDL receptor in the liver as well as by reducing the release of LDL into the circulation. As a monotherapy, the statin class of lipid lowering agents can reduce plasma LDL concentrations by 30-60% and triglycerides by 25%, producing a reduction in the incidence of CAD by 25-60% and the risk of death by 30%. Statins do not have an appreciable effect on HDL. A mechanistically distinct agent, Ezetimibe (Zetia, Merck and Co.), also possesses the ability to reduce plasma LDL, however it functions by inhibiting the absorption of cholesterol by the small intestine via antagonism of the NPC1L1 receptor (PNAS 2005, 102: 8132-8137). Monotherapy with Ezetimibe typically lowers LDL by 20%, however when co-formulated with a statin, maximal reductions can exceed 60%. As with the statins, however, Ezetimibe has a negligible effect on plasma HDL.
While statins can have a modest impact on circulating triglycerides, PPAR alpha agonists (or fibrates) are far superior in targeting this lipid endpoint. The fibrates function by increasing lipolysis and elimination of triglyceride-rich particles from plasma by activating lipoprotein lipase and reducing production of apolipoprotein C-III (an inhibitor of lipoprotein lipase activity). One such fibrate, Fenofibrate (Tricor, Abott), has been shown in clinical studies to decrease plasma triglyceride levels upwards of 40-60%. Interestingly, the fibrate class of lipid-lowering drugs also has a modest, but significant effect on both LDL (20% reduction) and HDL (10% increase).
Currently, the statin class of LDL lowering agents remains the cornerstone of dyslipidemia therapy. Despite the substantial reduction in cardiovascular events that have been achieved with this therapeutic approach, however, the cardio-protection that is afforded to patients by these therapies is still incomplete. It is now clear that therapies that are targeted to increase HDL cholesterol are critical in terms of maximizing patient cardio-protection. The only therapy available to date that has the ability to effectively raise circulating levels of cardio-protective HDL and consequently improve the progression of atherosclerosis in CAD patients is nicotinic acid (niacin or vitamin B3). Nicotinic acid was first reported to modify lipoprotein profiles in 1955 (Altschul et al. Arch Biochem Biophys 1955, 54: 558-559). Its effects are the most broad-spectrum of any available therapy, effectively raising HDL levels (20-30%) as well as lowering circulating plasma LDL (16%) and triglycerides (38%). The clinical significance of this broad-spectrum activity has been revealed in multiple large clinical studies. In the most recent ARBITER 2 (Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol 2; Taylor et al. Circulation 2004, 110: 3512-3517) study, patients on statin therapy were randomized to either placebo or 1000 mg extended release (ER) niacin (Niaspan, Kos Pharmaceuticals). Patients receiving niacin exhibited a statistically significant decrease in carotid intima-media thickness, a validated surrogate cardiovascular end point. This study also revealed a significantly reduced rate of intima-media thickness progression in subjects without detectable insulin resistance. This study indicates the incomplete cardio-protection that is offered by statin therapy and substantiates the utility of nicotinic acid in reducing overall cardiac risk in low-HDL patients.
While nicotinic acid has been used clinically to modify lipid profiles for over four decades, the mechanism of action of the compound has remained largely obscure. It has long been known that acute nicotinic acid dosing results in a profound decrease in circulating free fatty acids (FFAs). This anti-lipolytic activity was first hypothesized in 1980 to be mediated by a membrane receptor linked to a decrease in intracellular cAMP (cyclic AMP, or cyclic adenosine monophosphate, or 3′-5′-cyclic adenosine monophosphate) levels (Aktories et al. FEBS Letters 1980, 115: 11-14). This hypothesis was later confirmed and the implied Gi/o GPCR-coupling was verified using pertussis toxin sensitivity studies (Aktories et al. FEBS Letters 1983, 156: 88-92). The identification of specific nicotinic acid binding sites on the surface of adipose and spleen cells confirmed the membrane hypothesis and refined, using modern-day techniques, the G-protein coupling of the receptor itself (Lorenzen et al. Mol Pharm 2001, 59: 349-357). This G-protein mediated, anti-lipolytic activity of nicotinic acid was used for two decades to identify and characterize nicotinic acid analogues in terms of their therapeutic potential. Finally, in 2003, two independent groups simultaneously published the cloning of an orphan Gi/o-coupled GPCR, HM74a (Wise et al. J Biol Chem 2003, 278: 9869-9874; Tunaru et al. Nat Med 2003, 9: 352-355), which binds to nicotinic acid with high affinity. As predicted, this receptor was shown to be expressed in adipose tissue and spleen, and binds to not only nicotinic acid, but also to the structurally related derivatives that had been previously shown to exhibit adipocyte anti-lipolytic activity. Mice that have been made deficient in the rodent ortholog of HM74a (Puma-g) by homologous recombination resist nicotinic acid-dependent FFA reduction and TG lowering. It is currently hypothesized that the nicotinic acid anti-lipolytic activity is based on the activation of this high affinity GPCR (HM74a), resulting in a decrease in intracellular cAMP and a subsequent attenuation of hormone sensitive lipase (HSL) activity. Decreased adipocyte lipolytic output results in a reduction in circulating FFA and a corresponding reduction in hepatic TGs, very-low density LDL (VLDL), and LDL. The increased levels of HDL arise from an effective reduction of cholesterol ester transfer protein activity due to decreased availability of VLDL acceptor molecules.
Beyond impacting lipid levels and lipoprotein profiles, FFAs play fundamental roles in the regulation of glycemic control. It is now recognized that chronically elevated plasma FFA concentrations cause insulin resistance in muscle and liver, and impair insulin secretion (reviewed in Defronzo et al. Int. J. Clin. Prac. 2004, 58: 9-21). In muscle, acute elevations in plasma FFA concentrations can increase intramyocellular lipid content; this can have direct negative effects on insulin receptor signaling and glucose transport. In liver, increased plasma FFAs lead to accelerated lipid oxidation and acetyl-CoA accumulation, the later of which stimulates the rate-limiting steps for hepatic glucose production. In the pancreas, long-term exposure to elevated FFAs has been shown to impair the beta-cell's ability to secrete insulin in response to glucose. This data has driven the hypothesis that adipose tissue FFA release is a primary driver of the underlying pathologies in type 2 diabetes, and strategies designed to reduce FFAs, for example by agonizing HM74A, may prove effective in improving insulin sensitivity and lowering blood glucose levels in patients with type 2 diabetics/metabolic syndrome
The utility of nicotinic acid as a hypolipidemic/FFA lowering agent is currently limited by four main factors. First, significant doses of nicotinic acid are required to impact FFA release and improve lipid parameters. Immediate release (IR) nicotinic acid is often dosed at 3-9 g/day in order to achieve efficacy, and ER nicotinic acid (Niaspan) is typically dosed between 1-2 g/day. These high doses drive the second issue with nicotinic acid therapy, hepatotoxicity. One of the main metabolic routes for nicotinic acid is the formation of nicotinamide (NAM). Increased levels of NAM have been associated with elevated liver transaminase which can lead to hepatic dysfunction. This toxicity is particularly problematic for sustained release formulations and results in the need to monitor liver enzymes during the initiation of therapy. Third, high doses of nicotinic acid are associated with severe prostaglandin-mediated cutaneous flushing. Virtually all patients experience flushing when on IR-nicotinic acid at or near the Tmax of the drug and discontinuation of therapy occurs in 20-50% of individuals. Niaspan, while exhibiting an increased dissolution time, still possesses a flushing frequency of approximately 70%, and this is in spite of the recommended dosing regimen that includes taking Niaspan along with an aspirin after a low-fat snack. Fourth, nicotinic acid therapy often results in FFA rebound, a condition whereby free fatty acid levels are not adequately suppressed throughout the dosing regimen, resulting in a compensatory increase in adipose tissue lipolysis. With immediate release nicotinic acid therapy, this rebound phenomenon is so great that daily FFA AUCs are actually increased after therapy. Such FFA excursions can lead to impaired glycemic control and elevated blood glucose levels, both of which have been shown to occur in some individuals after nicotinic acid therapy.
Giving the importance of nicotinic acid in modulating (especially agonizing) HM74a receptor and its limitations, novel small molecules designed to mimic the mechanism of nicotinic acid's action on HM74a offer the possibility of achieving greater HDL, LDL, TG, and FFA efficacy while avoiding adverse effects such as hepatotoxicity and cutaneous flushing. Such therapies are envisoned to have significant impact beyond dyslipidemia to include insulin resistance, hyperglycemia, and associated syndromes by virtue of their ability to more adequately reduce plasma FFA levels during the dosing interval. The present invention is directed to these, as well as other, important ends.